Coherency of replicas for a distributed file sharing system

• US 7,831,735 B1
• Filed: 03/02/2009
• Issued: 11/09/2010
• Est. Priority Date: 12/31/2003
• Status: Expired due to Fees

• Current Assignee: Veritas Technologies, LLC (Whitehouse Group Ltd.)
• Original Assignee: Symantec Operating Corporation (Gen Digital Inc.)
• Inventors: Gunda, Laxmikant Vithal, Venkata, Sivaramakrishna Ramadugu, Ranade, Dilip Madhusudan, Banerjee, Anindya, Kabra, Navin, Nanda, BijayaLaxmi, Shelat, Radha
• Primary Examiner(s): Mesfin; Yemane
• Application Number:
  US12/396,280
• Time in Patent Office: 617 Days
• Field of Search: 709/203, 709/246, 709/248, 707/624, 707/694, 707/695
• US Class Current: 709/248
• CPC Class Codes: G06F 16/176 Support for shared access to files; File sharing support
• CPC Class Codes: G06F 16/1787 Details of non-transparently synchronising file systems
• CPC Class Codes: G06F 16/184 implemented as replicated file system
• CPC Class Codes: G06F 16/1865 Transactional file systems
• CPC Class Codes: Y10S 707/99952 Coherency, e.g. same view to multiple users

Abstract

A plurality of data objects may be replicated across a plurality of computing nodes coupled to a network. The network may include a first node operable to initiate an update operation to update a plurality of replicas of a first object. If one or more of the replicas are not reachable then the update operation may update a subset (e.g., a quorum) but not all of the replicas. For each node on which one of the replicas was updated in the update operation, the node may add the object to a list of incoherent objects. The list of incoherent objects may subsequently be used to bring the lagging replicas in sync with the replicas that were updated. In another embodiment, a plurality of replicas of an object may be stored on a plurality of nodes, similarly as described above. A first node that stores a replica of the object may store a first timestamp associated with the replica on the first node. The timestamp may be used to ensure that the replica on the first node is coherent with respect to one or more other replicas by periodically communicating with the one or more other replicas when a threshold amount of time has passed without the replica on the first node receiving an update.

Claims

1. A system comprising:

a plurality of computing nodes, including a first node configured to store a first timestamp associated with a first replica of an object, wherein the first replica is stored on the first node;

wherein in response to receiving a read request to read data from the object, the first node is configured to;

determine whether an amount of time elapsed since a time indicated by the first timestamp exceeds a threshold amount;

if the amount of time elapsed does not exceed the threshold amount, respond to the read request by returning data from the first replica stored on the first node;

if the amount of time elapsed does exceed the threshold amount, synchronize the first replica of the object stored on the first node with a second replica of the object stored on a second node of the plurality of nodes; and

after said synchronizing, respond to the read request by returning data from the first replica stored on the first node.

2. The system of claim 1, wherein the first node is configured to synchronize the first replica of the object stored on the first node with the second replica stored on the second node by:

communicating with the second node to determine whether the second replica stored on the second node has received one or more updates since the first replica stored on the first node was last known to be coherent with respect to the second replica stored on the second node; and

performing one of;

if the second replica has received one or more updates, update the first replica stored on the first node by applying the one or more updates to the first replica;

if the second replica has not received one or more updates, determine that the first replica stored on the first node does not need to be updated.

3. The system of claim 1,

wherein the first node is further configured to update the first timestamp associated with the first replica in response to synchronizing the first replica with the second replica.

4. The system of claim 3,

wherein the first node is configured to update the first timestamp by setting the first timestamp to specify the time at which the first replica was synchronized with the second replica.

5. The system of claim 3,

wherein the first node is configured to update the first timestamp associated with the first replica even if the second replica stored on the second node has not received one or more updates.

6. The system of claim 1,

wherein the first node is further configured to:

receive an update to the first replica of the object stored on the first node;

apply the update to the first replica of the object; and

update the first timestamp associated with the first replica in response to the update.

7. A non-transitory computer-accessible storage medium storing program instructions executable by a first node of a plurality of computing nodes to cause the first node to:

store a first replica of an object;

store a first timestamp associated with the first replica;

in response to receiving a read request to read data from the object;

determine whether an amount of time elapsed since a time indicated by the first timestamp exceeds a threshold amount;

if the amount of time elapsed does not exceed the threshold amount, respond to the read request by returning data from the first replica;

if the amount of time elapsed does exceed the threshold amount, synchronize the first replica of the object with a second replica of the object stored on a second node of the plurality of nodes; and

after said synchronizing, respond to the read request by returning data from the first replica of the object.

8. The computer-accessible storage medium of claim 7, wherein the program instructions are executable to cause the first node to synchronize the first replica of the object with the second replica stored on the second node by:

communicating with the second node to determine whether the second replica stored on the second node has received one or more updates since the first replica was last known to be coherent with respect to the second replica stored on the second node; and

performing one of;

if the second replica has received one or more updates, update the first replica by applying the one or more updates to the first replica;

if the second replica has not received one or more updates, determine that the first replica does not need to be updated.

9. The computer-accessible storage medium of claim 7,

wherein the program instructions are further executable to cause the first node to update the first timestamp associated with the first replica in response to synchronizing the first replica with the second replica.

10. The computer-accessible storage medium of claim 9,

wherein the program instructions are executable to cause the first node to update the first timestamp by setting the first timestamp to specify the time at which the first replica was synchronized with the second replica.

11. The computer-accessible storage medium of claim 9,

wherein the program instructions are executable to cause the first node to update the first timestamp associated with the first replica even if the second replica stored on the second node has not received one or more updates.

12. The computer-accessible storage medium of claim 7, wherein the program instructions are further executable to cause the first node to:

receive an update to the first replica of the object;

apply the update to the first replica of the object; and

update the first timestamp associated with the first replica in response to the update.

13. A method comprising:

a first node of a plurality of computing nodes storing a first replica of an object;

the first node storing a first timestamp associated with the first replica;

in response to the first node receiving a read request to read data from the object;

the first node determining whether an amount of time elapsed since a time indicated by the first timestamp exceeds a threshold amount; and

if the amount of time elapsed does not exceed the threshold amount, the first node responding to the read request by returning data from the first replica stored on the first node;

if the amount of time elapsed does exceed the threshold amount, the first node synchronizing the first replica of the object stored on the first node with a second replica of the object stored on a second node of the plurality of nodes; and

after said synchronizing, the first node responding to the read request by returning data from the first replica stored on the first node.

14. The method of claim 13, wherein the first node synchronizes the first replica of the object stored on the first node with the second replica stored on the second node by:

communicating with the second node to determine whether the second replica stored on the second node has received one or more updates since the first replica stored on the first node was last known to be coherent with respect to the second replica stored on the second node; and

performing one of;

if the second replica has received one or more updates, updating the first replica stored on the first node by applying the one or more updates to the first replica;

if the second replica has not received one or more updates, determining that the first replica stored on the first node does not need to be updated.